The general aim of the proposed research is to contribute empirically and theoretically to an understanding of aversively motivated learning. Because the two-factor theory of avoidance learning cannot readily account for certain data, it is proposed to modify it by incorporating into the theory the concepts of conditioned relaxation and effective reinforcement. The proposed experiments are aimed at determining the feasibility of making such modifications. Some of the research will be concerned with the manner in which relaxation becomes conditioned to neutral stimuli and how relaxation serves to reinforce avoidance behavior by reducing fear. These studies will contrast the relaxation theory with an alternative position which holds that such stimuli become conditioned inhibitors. Other studies will test implications of the effective reinforcement theory which holds that reinforcement depends not only on the amount of fear reduction occurring when an instrumental response is made but also on the amount of fear remaining after the response.